Improving the Intelligent Classification of Crease Patterns in OSELM Fabric Folding
In order to improve the accuracy of garment fabric crease classification,a DE-SCA-OSELM garment fabric crease classification method is proposed based on the principle of online sequential limit learning machine.Firstly,OSELM was used as the basic classification model.Then,the detection peak operator in the differential evolution algorithm DE was employed to initialize the poor individuals in the sine and cosine optimization algorithm SCA,so as to increase the population diversity and improve the global optimization ability of the SCA algorithm.Finally,the DE-SCA algorithm was utilized to optimize the parameters of the OSELM classification model to achieve accurate crease classification of garment fab-ric.The simulation results show that under the condition of optimal parameters,the fabric crease classifi-cation accuracy can reach 94.17％,which is higher than other traditional crease classification algorithms,indicating that DE-SCA-OSELM enjoys good classification performance and can lay a foundation for sub-sequent crease classification.
万方数据
维普
